11 Mar 2004
A LIDAR system is helping US authorities monitor the number of salmon returning to their birthplace.
Scientists in the US have packed a LIDAR system into a small aircraft and used it to monitor adult salmon returning to the stream where they were born. Built around a pulsed Nd:YAG laser and a gated intensified CCD camera, the developers from Colorado say their airborne LIDAR provides information on the number and location of the fish. (Applied Optics 43 1416)
“Salmon returning to spawn are fished by commercial, sport and subsistance fishers as well as natural predators,” explained one of the paper’s authors James Churnside from the National Oceanic and Atmospheric Administration. “Management involves opening and closing different areas to fishing to ensure that enough fish survive to spawn.”
According to Churnside, airborne LIDAR is ideal as large areas can be mapped quickly which gives fishery managers an overall snapshot of the distribution of the fish. The researchers fly the system over the area of interest at a speed of 65 meters per second and an altitude of 150 meters.
The LIDAR system used by Churnside and his colleague James Wilson is based on a Q-switched Nd:YAG laser. It produces 12 mJ of 532-nm light in a 12 ns pulse at a repetition rate of 30 Hz. They also use a clever mixture of electronics and a gated intensified CCD to image the fish from the aircraft.
While the laser can penetrate some 50 meters into the water, the duo admits that their camera system can only image fish to a depth of a couple of meters. Luckily, salmon returning to their natal stream tend to swim close to the water’s surface.
The resulting images show an area of approximately 3.75 x 2.81 meters. According to Churnside, the spatial resolution of the images is 5.9 mm which is sufficient to resolve individual fish.
Churnside and Wilson also process their images so that the fish can be counted automatically. “We are considering the use of the imaging system for identification of schooling fish like sardines by their school shape,” said Churnside. “These schools are generally deeper however and we need to increase the depth of penetration of the imaging system for that to be successful. We are looking at ways to increase the signal to the camera.”